HOWTO : CUDA with Kali Linux 1.0.9

The guide has been updated on FEB 4, 2015 as Kali Linux 1.0.9a includes the nVidia driver 340.x and CUDA 5.5.x.

Hardware

CPU : Intel i7-3930K
RAM : 32GB DDR3
Hard Drive : 3TB
Display Card : Two nVidia GeForce GTX 590

Install Kali

Install Kali Linux 1.0.9 on the box as usual. Make sure "secure boot" is disabled in your BIOS before installing. After that, you update the Kali accordingly.

apt-get update
apt-get dist-upgrade

Install nVidia Driver

apt-get install -y linux-headers-$(uname -r)
apt-get install nvidia-kernel-dkms nvidia-driver nvidia-cuda-toolkit nvidia-xconfig

nvidia-xconfig

sed 's/quiet/quiet nouveau.modeset=0/g' -i /etc/default/grub
update-grub
reboot

Fix the nvidia_uvm error

After the reboot :

cd /usr/src/nvidia-current-331.67
make

cp Module.symvers uvm/
make -C uvm

cp uvm/nvidia-uvm.ko /lib/modules/`uname -r`/updates/dkms

Updated on Jan 8, 2015

Remark : If you use backports version, the nvidia_uvm problem is fixed and the cudaHashcat 1.31 can be running without any problem. The backports version of nVidia driver is 340.65 and the version of CUDA driver is 5.5.22 at this time of writing.

echo "deb http://ftp.debian.org/debian wheezy-backports main contrib non-free" >> /etc/apt/sources.list

apt-get update

apt-get install -t wheezy-backports nvidia-kernel-dkms nvidia-cuda-toolkit nvidia-driver nvidia-xconfig

If you have done the following steps, you are not required to re-do it again.

nvidia-xconfig

sed 's/quiet/quiet nouveau.modeset=0/g' -i /etc/default/grub
update-grub
reboot

Install cudaHashcat

mkdir hacking
cd hacking

wget http://hashcat.net/files/cudaHashcat-1.30.7z

7za x cudaHashcat-1.30.7z

wget http://hashcat.net/files-legacy/cudaHashcat-1.31.7z

7za x cudaHashcat-1.31.7z

(Please noted that the current version 1.32 does not compatible to nVidia driver 340.x).

Test the cudaHashcat

cd /root/hacking/cudaHashcat-1.30/
./cudaExample0.sh

cd /root/hacking/cudaHashcat-1.30/
./cudaExample400.sh

cd /root/hacking/cudaHashcat-1.30/
./cudaExample500.sh

Install John the Ripper

(Please note that the current version of john is john-1.80-jumbo-1.tar.gz)

apt-get install libssl-dev

cd hacking

wget http://www.openwall.com/john/g/john-1.7.9-jumbo-7.tar.gz
tar -xvzf john-1.7.9-jumbo-7.tar.gz
cd john-1.7.9-jumbo-7/src

./configure
make
make clean linux-x86-64-cuda

** If your hashes or passwords are longer than 8 characters, you need to change the following before compiling the John.

cd john-1.7.9-jumbo-7/src
nano params.h

Then change from "8" to "18" or "20" and etc.

#define CHARSET_LENGTH 8

Test the John the Ripper

cd /root/hacking/john-1.7.9-jumbo-7/run

./john --device=0,1,2,3 --format=sha512crypt-cuda /etc/shadow

* since I have 4 GPUs, so the --device should be 4.

*** When you changed the CHARSET_LENGTH, you need to generate a new charset. Do it once only.

wget http://downloads.skullsecurity.org/passwords/rockyou.txt.bz2
bunzip2 -d rockyou.txt.bz2
cp rockyou.txt /root/hacking/john-1.7.9-jumbo-7/run

cd /root/hacking/john-1.7.9-jumbo-7/run

cat rockyou.txt | sed 's/^/:/' > rockyou.pot

mv all.chr all.chr-original
mv alnum.chr alnum.chr-original
mv alpha.chr alpha.chr-original
mv digits.chr digits.chr-original
mv lanman.chr lanman.chr-original

./john --pot=rockyou.pot --make-charset=all.chr
./john --pot=rockyou.pot --make-charset=alnum.chr --external=filter_alnum
./john --pot=rockyou.pot --make-charset=alpha.chr --external=filter_alpha
./john --pot=rockyou.pot --make-charset=digits.chr --external=filter_digits
./john --pot=rockyou.pot --make-charset=lanman.chr --external=filter_lanman

Then your cracking command will be :

./john --pot=rockyou.pot --device=0,1,2,3 --format=sha512crypt-cuda /etc/shadow

Install and Test Cryptohaze

cd hacking
wget http://sourceforge.net/projects/cryptohaze/files/Cryptohaze-Linux_x64_1_31a.tar.bz2/download -O Cryptohaze-Linux_x64_1_31a.tar.bz2

tar xjvf Cryptohaze-Linux_x64_1_31a.tar.bz2

cd /root/hacking/Cryptohaze-Linux
./Cryptohaze-Multiforcer -h NTLM -c charsets/charsetall -f test_hashes/Hashes-NTLM-Full.txt

That's all! See you.